CyPhyHouse aims to provide programming, debugging, and deployment benefits for distributed mobile robotic applications. Develop hardware-agnostic, distributed applications using the high-level, event driven Koord programming language included
with CyPhyHouse, without requiring expertise in controller design or distributed network protocols.
Our high level programming language, Koord, makes code clean and succinct through abstractions.
We use Python 3 to implement the CyPhyHouse middleware, with a modular and extensible design for greater flexibility.
Our simulation environment in Gazebo allows convenient testing and debugging of multi-robot applications.
We are working on Koord enabled formal verification of multi-robot applications using a variety of validation tools.
Control and actuation in the CyPhyHouse toolchain are done in simulation and deployment using ROS.
We have deployed several Koord applications in the Flight Arena (equipped with VICON) at the IRL at UIUC.
Talk at Workshop on Safe Autonomy: Learning, Verification, and Trusted Operation of Autonomous Systems with c3.ai
A quick overview of CyPhyHouse, including the design of a distributed task allocation application
Talk at Workshop on Safe Operation of Connected and Autonomous Vehicle Fleets
Video presentation for OOPSLA 2020 on Koord programming language and verification
Student researchers interacting with the local community about CyPhyHouse.
A successful demo at the EOH 2018 of a firefighting application.
Celebrating a productive summer with some of our talented and hard-working undergraduate team members.
Simulation video of drone formation flight in our simulation environment in Gazebo. Drones automatically align themselves with their neighbors through communication except for the drones at the corners.
Simulation video of distributed task allocation in our simulation environment. The grey circles are unassigned tasks, red are assigned, and green are completed.
A mapping application simulated in the Gazebo simulation environment, where two robots collaboratively build an approximate map of unknown static obstacles using LIDAR sensing.